rts/PrimOps.cmm

   1 /* -----------------------------------------------------------------------------
   2  *
   3  * (c) The GHC Team, 1998-2004
   4  *
   5  * Out-of-line primitive operations
   6  *
   7  * This file contains the implementations of all the primitive
   8  * operations ("primops") which are not expanded inline.  See
   9  * ghc/compiler/prelude/primops.txt.pp for a list of all the primops;
  10  * this file contains code for most of those with the attribute
  11  * out_of_line=True.
  12  *
  13  * Entry convention: the entry convention for a primop is that all the
  14  * args are in Stg registers (R1, R2, etc.).  This is to make writing
  15  * the primops easier.  (see compiler/codeGen/CgCallConv.hs).
  16  *
  17  * Return convention: results from a primop are generally returned
  18  * using the ordinary unboxed tuple return convention.  The C-- parser
  19  * implements the RET_xxxx() macros to perform unboxed-tuple returns
  20  * based on the prevailing return convention.
  21  *
  22  * This file is written in a subset of C--, extended with various
  23  * features specific to GHC.  It is compiled by GHC directly.  For the
  24  * syntax of .cmm files, see the parser in ghc/compiler/cmm/CmmParse.y.
  25  *
  26  * ---------------------------------------------------------------------------*/
  27
  28 #include "Cmm.h"
  29
  30 #ifdef __PIC__
  31 #ifndef mingw32_HOST_OS
  32 import __gmpz_init;
  33 import __gmpz_add;
  34 import __gmpz_sub;
  35 import __gmpz_mul;
  36 import __gmpz_gcd;
  37 import __gmpn_gcd_1;
  38 import __gmpn_cmp;
  39 import __gmpz_tdiv_q;
  40 import __gmpz_tdiv_r;
  41 import __gmpz_tdiv_qr;
  42 import __gmpz_fdiv_qr;
  43 import __gmpz_divexact;
  44 import __gmpz_and;
  45 import __gmpz_xor;
  46 import __gmpz_ior;
  47 import __gmpz_com;
  48 #endif
  49 import pthread_mutex_lock;
  50 import pthread_mutex_unlock;
  51 #endif
  52 import base_GHCziIOBase_NestedAtomically_closure;
  53 import EnterCriticalSection;
  54 import LeaveCriticalSection;
  55
  56 /*-----------------------------------------------------------------------------
  57   Array Primitives
  58
  59   Basically just new*Array - the others are all inline macros.
  60
  61   The size arg is always passed in R1, and the result returned in R1.
  62
  63   The slow entry point is for returning from a heap check, the saved
  64   size argument must be re-loaded from the stack.
  65   -------------------------------------------------------------------------- */
  66
  67 /* for objects that are *less* than the size of a word, make sure we
  68  * round up to the nearest word for the size of the array.
  69  */
  70
  71 newByteArrayzh_fast
  72 {
  73     W_ words, payload_words, n, p;
  74     MAYBE_GC(NO_PTRS,newByteArrayzh_fast);
  75     n = R1;
  76     payload_words = ROUNDUP_BYTES_TO_WDS(n);
  77     words = BYTES_TO_WDS(SIZEOF_StgArrWords) + payload_words;
  78     ("ptr" p) = foreign "C" allocateLocal(MyCapability() "ptr",words) [];
  79     TICK_ALLOC_PRIM(SIZEOF_StgArrWords,WDS(payload_words),0);
  80     SET_HDR(p, stg_ARR_WORDS_info, W_[CCCS]);
  81     StgArrWords_words(p) = payload_words;
  82     RET_P(p);
  83 }
  84
  85 newPinnedByteArrayzh_fast
  86 {
  87     W_ words, payload_words, n, p;
  88
  89     MAYBE_GC(NO_PTRS,newPinnedByteArrayzh_fast);
  90     n = R1;
  91     payload_words = ROUNDUP_BYTES_TO_WDS(n);
  92
  93     // We want an 8-byte aligned array.  allocatePinned() gives us
  94     // 8-byte aligned memory by default, but we want to align the
  95     // *goods* inside the ArrWords object, so we have to check the
  96     // size of the ArrWords header and adjust our size accordingly.
  97     words = BYTES_TO_WDS(SIZEOF_StgArrWords) + payload_words;
  98     if ((SIZEOF_StgArrWords & 7) != 0) {
  99         words = words + 1;
 100     }
 101
 102     ("ptr" p) = foreign "C" allocatePinned(words) [];
 103     TICK_ALLOC_PRIM(SIZEOF_StgArrWords,WDS(payload_words),0);
 104
 105     // Again, if the ArrWords header isn't a multiple of 8 bytes, we
 106     // have to push the object forward one word so that the goods
 107     // fall on an 8-byte boundary.
 108     if ((SIZEOF_StgArrWords & 7) != 0) {
 109         p = p + WDS(1);
 110     }
 111
 112     SET_HDR(p, stg_ARR_WORDS_info, W_[CCCS]);
 113     StgArrWords_words(p) = payload_words;
 114     RET_P(p);
 115 }
 116
 117 newArrayzh_fast
 118 {
 119     W_ words, n, init, arr, p;
 120     /* Args: R1 = words, R2 = initialisation value */
 121
 122     n = R1;
 123     MAYBE_GC(R2_PTR,newArrayzh_fast);
 124
 125     words = BYTES_TO_WDS(SIZEOF_StgMutArrPtrs) + n;
 126     ("ptr" arr) = foreign "C" allocateLocal(MyCapability() "ptr",words) [R2];
 127     TICK_ALLOC_PRIM(SIZEOF_StgMutArrPtrs, WDS(n), 0);
 128
 129     SET_HDR(arr, stg_MUT_ARR_PTRS_DIRTY_info, W_[CCCS]);
 130     StgMutArrPtrs_ptrs(arr) = n;
 131
 132     // Initialise all elements of the the array with the value in R2
 133     init = R2;
 134     p = arr + SIZEOF_StgMutArrPtrs;
 135   for:
 136     if (p < arr + WDS(words)) {
 137         W_[p] = init;
 138         p = p + WDS(1);
 139         goto for;
 140     }
 141
 142     RET_P(arr);
 143 }
 144
 145 unsafeThawArrayzh_fast
 146 {
 147   // SUBTLETY TO DO WITH THE OLD GEN MUTABLE LIST
 148   //
 149   // A MUT_ARR_PTRS lives on the mutable list, but a MUT_ARR_PTRS_FROZEN
 150   // normally doesn't.  However, when we freeze a MUT_ARR_PTRS, we leave
 151   // it on the mutable list for the GC to remove (removing something from
 152   // the mutable list is not easy, because the mut_list is only singly-linked).
 153   //
 154   // So that we can tell whether a MUT_ARR_PTRS_FROZEN is on the mutable list,
 155   // when we freeze it we set the info ptr to be MUT_ARR_PTRS_FROZEN0
 156   // to indicate that it is still on the mutable list.
 157   //
 158   // So, when we thaw a MUT_ARR_PTRS_FROZEN, we must cope with two cases:
 159   // either it is on a mut_list, or it isn't.  We adopt the convention that
 160   // the closure type is MUT_ARR_PTRS_FROZEN0 if it is on the mutable list,
 161   // and MUT_ARR_PTRS_FROZEN otherwise.  In fact it wouldn't matter if
 162   // we put it on the mutable list more than once, but it would get scavenged
 163   // multiple times during GC, which would be unnecessarily slow.
 164   //
 165   if (StgHeader_info(R1) != stg_MUT_ARR_PTRS_FROZEN0_info) {
 166         SET_INFO(R1,stg_MUT_ARR_PTRS_DIRTY_info);
 167         recordMutable(R1, R1);
 168         // must be done after SET_INFO, because it ASSERTs closure_MUTABLE()
 169         RET_P(R1);
 170   } else {
 171         SET_INFO(R1,stg_MUT_ARR_PTRS_DIRTY_info);
 172         RET_P(R1);
 173   }
 174 }
 175
 176 /* -----------------------------------------------------------------------------
 177    MutVar primitives
 178    -------------------------------------------------------------------------- */
 179
 180 newMutVarzh_fast
 181 {
 182     W_ mv;
 183     /* Args: R1 = initialisation value */
 184
 185     ALLOC_PRIM( SIZEOF_StgMutVar, R1_PTR, newMutVarzh_fast);
 186
 187     mv = Hp - SIZEOF_StgMutVar + WDS(1);
 188     SET_HDR(mv,stg_MUT_VAR_DIRTY_info,W_[CCCS]);
 189     StgMutVar_var(mv) = R1;
 190
 191     RET_P(mv);
 192 }
 193
 194 atomicModifyMutVarzh_fast
 195 {
 196     W_ mv, z, x, y, r;
 197     /* Args: R1 :: MutVar#,  R2 :: a -> (a,b) */
 198
 199     /* If x is the current contents of the MutVar#, then
 200        We want to make the new contents point to
 201
 202          (sel_0 (f x))
 203
 204        and the return value is
 205
 206          (sel_1 (f x))
 207
 208         obviously we can share (f x).
 209
 210          z = [stg_ap_2 f x]  (max (HS + 2) MIN_UPD_SIZE)
 211          y = [stg_sel_0 z]   (max (HS + 1) MIN_UPD_SIZE)
 212          r = [stg_sel_1 z]   (max (HS + 1) MIN_UPD_SIZE)
 213     */
 214
 215 #if MIN_UPD_SIZE > 1
 216 #define THUNK_1_SIZE (SIZEOF_StgThunkHeader + WDS(MIN_UPD_SIZE))
 217 #define TICK_ALLOC_THUNK_1() TICK_ALLOC_UP_THK(WDS(1),WDS(MIN_UPD_SIZE-1))
 218 #else
 219 #define THUNK_1_SIZE (SIZEOF_StgThunkHeader + WDS(1))
 220 #define TICK_ALLOC_THUNK_1() TICK_ALLOC_UP_THK(WDS(1),0)
 221 #endif
 222
 223 #if MIN_UPD_SIZE > 2
 224 #define THUNK_2_SIZE (SIZEOF_StgThunkHeader + WDS(MIN_UPD_SIZE))
 225 #define TICK_ALLOC_THUNK_2() TICK_ALLOC_UP_THK(WDS(2),WDS(MIN_UPD_SIZE-2))
 226 #else
 227 #define THUNK_2_SIZE (SIZEOF_StgThunkHeader + WDS(2))
 228 #define TICK_ALLOC_THUNK_2() TICK_ALLOC_UP_THK(WDS(2),0)
 229 #endif
 230
 231 #define SIZE (THUNK_2_SIZE + THUNK_1_SIZE + THUNK_1_SIZE)
 232
 233    HP_CHK_GEN_TICKY(SIZE, R1_PTR & R2_PTR, atomicModifyMutVarzh_fast);
 234
 235 #if defined(THREADED_RTS)
 236     ACQUIRE_LOCK(atomic_modify_mutvar_mutex "ptr") [R1,R2];
 237 #endif
 238
 239    x = StgMutVar_var(R1);
 240
 241    TICK_ALLOC_THUNK_2();
 242    CCCS_ALLOC(THUNK_2_SIZE);
 243    z = Hp - THUNK_2_SIZE + WDS(1);
 244    SET_HDR(z, stg_ap_2_upd_info, W_[CCCS]);
 245    LDV_RECORD_CREATE(z);
 246    StgThunk_payload(z,0) = R2;
 247    StgThunk_payload(z,1) = x;
 248
 249    TICK_ALLOC_THUNK_1();
 250    CCCS_ALLOC(THUNK_1_SIZE);
 251    y = z - THUNK_1_SIZE;
 252    SET_HDR(y, stg_sel_0_upd_info, W_[CCCS]);
 253    LDV_RECORD_CREATE(y);
 254    StgThunk_payload(y,0) = z;
 255
 256    StgMutVar_var(R1) = y;
 257    foreign "C" dirty_MUT_VAR(BaseReg "ptr", R1 "ptr") [R1];
 258
 259    TICK_ALLOC_THUNK_1();
 260    CCCS_ALLOC(THUNK_1_SIZE);
 261    r = y - THUNK_1_SIZE;
 262    SET_HDR(r, stg_sel_1_upd_info, W_[CCCS]);
 263    LDV_RECORD_CREATE(r);
 264    StgThunk_payload(r,0) = z;
 265
 266 #if defined(THREADED_RTS)
 267     RELEASE_LOCK(atomic_modify_mutvar_mutex "ptr") [];
 268 #endif
 269
 270    RET_P(r);
 271 }
 272
 273 /* -----------------------------------------------------------------------------
 274    Weak Pointer Primitives
 275    -------------------------------------------------------------------------- */
 276
 277 STRING(stg_weak_msg,"New weak pointer at %p\n")
 278
 279 mkWeakzh_fast
 280 {
 281   /* R1 = key
 282      R2 = value
 283      R3 = finalizer (or NULL)
 284   */
 285   W_ w;
 286
 287   if (R3 == NULL) {
 288     R3 = stg_NO_FINALIZER_closure;
 289   }
 290
 291   ALLOC_PRIM( SIZEOF_StgWeak, R1_PTR & R2_PTR & R3_PTR, mkWeakzh_fast );
 292
 293   w = Hp - SIZEOF_StgWeak + WDS(1);
 294   SET_HDR(w, stg_WEAK_info, W_[CCCS]);
 295
 296   StgWeak_key(w)       = R1;
 297   StgWeak_value(w)     = R2;
 298   StgWeak_finalizer(w) = R3;
 299
 300   StgWeak_link(w)       = W_[weak_ptr_list];
 301   W_[weak_ptr_list]     = w;
 302
 303   IF_DEBUG(weak, foreign "C" debugBelch(stg_weak_msg,w) []);
 304
 305   RET_P(w);
 306 }
 307
 308
 309 finalizzeWeakzh_fast
 310 {
 311   /* R1 = weak ptr
 312    */
 313   W_ w, f;
 314
 315   w = R1;
 316
 317   // already dead?
 318   if (GET_INFO(w) == stg_DEAD_WEAK_info) {
 319       RET_NP(0,stg_NO_FINALIZER_closure);
 320   }
 321
 322   // kill it
 323 #ifdef PROFILING
 324   // @LDV profiling
 325   // A weak pointer is inherently used, so we do not need to call
 326   // LDV_recordDead_FILL_SLOP_DYNAMIC():
 327   //    LDV_recordDead_FILL_SLOP_DYNAMIC((StgClosure *)w);
 328   // or, LDV_recordDead():
 329   //    LDV_recordDead((StgClosure *)w, sizeofW(StgWeak) - sizeofW(StgProfHeader));
 330   // Furthermore, when PROFILING is turned on, dead weak pointers are exactly as
 331   // large as weak pointers, so there is no need to fill the slop, either.
 332   // See stg_DEAD_WEAK_info in StgMiscClosures.hc.
 333 #endif
 334
 335   //
 336   // Todo: maybe use SET_HDR() and remove LDV_recordCreate()?
 337   //
 338   SET_INFO(w,stg_DEAD_WEAK_info);
 339   LDV_RECORD_CREATE(w);
 340
 341   f = StgWeak_finalizer(w);
 342   StgDeadWeak_link(w) = StgWeak_link(w);
 343
 344   /* return the finalizer */
 345   if (f == stg_NO_FINALIZER_closure) {
 346       RET_NP(0,stg_NO_FINALIZER_closure);
 347   } else {
 348       RET_NP(1,f);
 349   }
 350 }
 351
 352 deRefWeakzh_fast
 353 {
 354   /* R1 = weak ptr */
 355   W_ w, code, val;
 356
 357   w = R1;
 358   if (GET_INFO(w) == stg_WEAK_info) {
 359     code = 1;
 360     val = StgWeak_value(w);
 361   } else {
 362     code = 0;
 363     val = w;
 364   }
 365   RET_NP(code,val);
 366 }
 367
 368 /* -----------------------------------------------------------------------------
 369    Arbitrary-precision Integer operations.
 370
 371    There are some assumptions in this code that mp_limb_t == W_.  This is
 372    the case for all the platforms that GHC supports, currently.
 373    -------------------------------------------------------------------------- */
 374
 375 int2Integerzh_fast
 376 {
 377    /* arguments: R1 = Int# */
 378
 379    W_ val, s, p;        /* to avoid aliasing */
 380
 381    val = R1;
 382    ALLOC_PRIM( SIZEOF_StgArrWords + WDS(1), NO_PTRS, int2Integerzh_fast );
 383
 384    p = Hp - SIZEOF_StgArrWords;
 385    SET_HDR(p, stg_ARR_WORDS_info, W_[CCCS]);
 386    StgArrWords_words(p) = 1;
 387
 388    /* mpz_set_si is inlined here, makes things simpler */
 389    if (%lt(val,0)) {
 390         s  = -1;
 391         Hp(0) = -val;
 392    } else {
 393      if (%gt(val,0)) {
 394         s = 1;
 395         Hp(0) = val;
 396      } else {
 397         s = 0;
 398      }
 399   }
 400
 401    /* returns (# size  :: Int#,
 402                  data  :: ByteArray#
 403                #)
 404    */
 405    RET_NP(s,p);
 406 }
 407
 408 word2Integerzh_fast
 409 {
 410    /* arguments: R1 = Word# */
 411
 412    W_ val, s, p;        /* to avoid aliasing */
 413
 414    val = R1;
 415
 416    ALLOC_PRIM( SIZEOF_StgArrWords + WDS(1), NO_PTRS, word2Integerzh_fast);
 417
 418    p = Hp - SIZEOF_StgArrWords;
 419    SET_HDR(p, stg_ARR_WORDS_info, W_[CCCS]);
 420    StgArrWords_words(p) = 1;
 421
 422    if (val != 0) {
 423         s = 1;
 424         W_[Hp] = val;
 425    } else {
 426         s = 0;
 427    }
 428
 429    /* returns (# size  :: Int#,
 430                  data  :: ByteArray# #)
 431    */
 432    RET_NP(s,p);
 433 }
 434
 435
 436 /*
 437  * 'long long' primops for converting to/from Integers.
 438  */
 439
 440 #ifdef SUPPORT_LONG_LONGS
 441
 442 int64ToIntegerzh_fast
 443 {
 444    /* arguments: L1 = Int64# */
 445
 446    L_ val;
 447    W_ hi, lo, s, neg, words_needed, p;
 448
 449    val = L1;
 450    neg = 0;
 451
 452    hi = TO_W_(val >> 32);
 453    lo = TO_W_(val);
 454
 455    if ( hi == 0 || (hi == 0xFFFFFFFF && lo != 0) )  {
 456        // minimum is one word
 457        words_needed = 1;
 458    } else {
 459        words_needed = 2;
 460    }
 461
 462    ALLOC_PRIM( SIZEOF_StgArrWords + WDS(words_needed),
 463                NO_PTRS, int64ToIntegerzh_fast );
 464
 465    p = Hp - SIZEOF_StgArrWords - WDS(words_needed) + WDS(1);
 466    SET_HDR(p, stg_ARR_WORDS_info, W_[CCCS]);
 467    StgArrWords_words(p) = words_needed;
 468
 469    if ( %lt(hi,0) ) {
 470      neg = 1;
 471      lo = -lo;
 472      if(lo == 0) {
 473        hi = -hi;
 474      } else {
 475        hi = -hi - 1;
 476      }
 477    }
 478
 479    if ( words_needed == 2 )  {
 480       s = 2;
 481       Hp(-1) = lo;
 482       Hp(0) = hi;
 483    } else {
 484        if ( lo != 0 ) {
 485            s = 1;
 486            Hp(0) = lo;
 487        } else /* val==0 */  {
 488            s = 0;
 489        }
 490    }
 491    if ( neg != 0 ) {
 492         s = -s;
 493    }
 494
 495    /* returns (# size  :: Int#,
 496                  data  :: ByteArray# #)
 497    */
 498    RET_NP(s,p);
 499 }
 500 word64ToIntegerzh_fast
 501 {
 502    /* arguments: L1 = Word64# */
 503
 504    L_ val;
 505    W_ hi, lo, s, words_needed, p;
 506
 507    val = L1;
 508    hi = TO_W_(val >> 32);
 509    lo = TO_W_(val);
 510
 511    if ( hi != 0 ) {
 512       words_needed = 2;
 513    } else {
 514       words_needed = 1;
 515    }
 516
 517    ALLOC_PRIM( SIZEOF_StgArrWords + WDS(words_needed),
 518                NO_PTRS, word64ToIntegerzh_fast );
 519
 520    p = Hp - SIZEOF_StgArrWords - WDS(words_needed) + WDS(1);
 521    SET_HDR(p, stg_ARR_WORDS_info, W_[CCCS]);
 522    StgArrWords_words(p) = words_needed;
 523
 524    if ( hi != 0 ) {
 525      s = 2;
 526      Hp(-1) = lo;
 527      Hp(0)  = hi;
 528    } else {
 529       if ( lo != 0 ) {
 530         s = 1;
 531         Hp(0) = lo;
 532      } else /* val==0 */  {
 533       s = 0;
 534      }
 535   }
 536
 537    /* returns (# size  :: Int#,
 538                  data  :: ByteArray# #)
 539    */
 540    RET_NP(s,p);
 541 }
 542
 543
 544
 545 #endif /* SUPPORT_LONG_LONGS */
 546
 547 /* ToDo: this is shockingly inefficient */
 548
 549 #ifndef THREADED_RTS
 550 section "bss" {
 551   mp_tmp1:
 552     bits8 [SIZEOF_MP_INT];
 553 }
 554
 555 section "bss" {
 556   mp_tmp2:
 557     bits8 [SIZEOF_MP_INT];
 558 }
 559
 560 section "bss" {
 561   mp_result1:
 562     bits8 [SIZEOF_MP_INT];
 563 }
 564
 565 section "bss" {
 566   mp_result2:
 567     bits8 [SIZEOF_MP_INT];
 568 }
 569 #endif
 570
 571 #ifdef THREADED_RTS
 572 #define FETCH_MP_TEMP(X) \
 573 W_ X; \
 574 X = BaseReg + (OFFSET_StgRegTable_r ## X);
 575 #else
 576 #define FETCH_MP_TEMP(X) /* Nothing */
 577 #endif
 578
 579 #define GMP_TAKE2_RET1(name,mp_fun)                                     \
 580 name                                                                    \
 581 {                                                                       \
 582   CInt s1, s2;                                                          \
 583   W_ d1, d2;                                                            \
 584   FETCH_MP_TEMP(mp_tmp1);                                               \
 585   FETCH_MP_TEMP(mp_tmp2);                                               \
 586   FETCH_MP_TEMP(mp_result1)                                             \
 587   FETCH_MP_TEMP(mp_result2);                                            \
 588                                                                         \
 589   /* call doYouWantToGC() */                                            \
 590   MAYBE_GC(R2_PTR & R4_PTR, name);                                      \
 591                                                                         \
 592   s1 = W_TO_INT(R1);                                                    \
 593   d1 = R2;                                                              \
 594   s2 = W_TO_INT(R3);                                                    \
 595   d2 = R4;                                                              \
 596                                                                         \
 597   MP_INT__mp_alloc(mp_tmp1) = W_TO_INT(StgArrWords_words(d1));          \
 598   MP_INT__mp_size(mp_tmp1)  = (s1);                                     \
 599   MP_INT__mp_d(mp_tmp1)     = BYTE_ARR_CTS(d1);                         \
 600   MP_INT__mp_alloc(mp_tmp2) = W_TO_INT(StgArrWords_words(d2));          \
 601   MP_INT__mp_size(mp_tmp2)  = (s2);                                     \
 602   MP_INT__mp_d(mp_tmp2)     = BYTE_ARR_CTS(d2);                         \
 603                                                                         \
 604   foreign "C" __gmpz_init(mp_result1 "ptr") [];                            \
 605                                                                         \
 606   /* Perform the operation */                                           \
 607   foreign "C" mp_fun(mp_result1 "ptr",mp_tmp1  "ptr",mp_tmp2  "ptr") []; \
 608                                                                         \
 609   RET_NP(TO_W_(MP_INT__mp_size(mp_result1)),                            \
 610          MP_INT__mp_d(mp_result1) - SIZEOF_StgArrWords);                \
 611 }
 612
 613 #define GMP_TAKE1_RET1(name,mp_fun)                                     \
 614 name                                                                    \
 615 {                                                                       \
 616   CInt s1;                                                              \
 617   W_ d1;                                                                \
 618   FETCH_MP_TEMP(mp_tmp1);                                               \
 619   FETCH_MP_TEMP(mp_result1)                                             \
 620                                                                         \
 621   /* call doYouWantToGC() */                                            \
 622   MAYBE_GC(R2_PTR, name);                                               \
 623                                                                         \
 624   d1 = R2;                                                              \
 625   s1 = W_TO_INT(R1);                                                    \
 626                                                                         \
 627   MP_INT__mp_alloc(mp_tmp1)     = W_TO_INT(StgArrWords_words(d1));      \
 628   MP_INT__mp_size(mp_tmp1)      = (s1);                                 \
 629   MP_INT__mp_d(mp_tmp1)         = BYTE_ARR_CTS(d1);                     \
 630                                                                         \
 631   foreign "C" __gmpz_init(mp_result1 "ptr") [];                            \
 632                                                                         \
 633   /* Perform the operation */                                           \
 634   foreign "C" mp_fun(mp_result1 "ptr",mp_tmp1 "ptr") [];                \
 635                                                                         \
 636   RET_NP(TO_W_(MP_INT__mp_size(mp_result1)),                            \
 637          MP_INT__mp_d(mp_result1) - SIZEOF_StgArrWords);                \
 638 }
 639
 640 #define GMP_TAKE2_RET2(name,mp_fun)                                                     \
 641 name                                                                                    \
 642 {                                                                                       \
 643   CInt s1, s2;                                                                          \
 644   W_ d1, d2;                                                                            \
 645   FETCH_MP_TEMP(mp_tmp1);                                                               \
 646   FETCH_MP_TEMP(mp_tmp2);                                                               \
 647   FETCH_MP_TEMP(mp_result1)                                                             \
 648   FETCH_MP_TEMP(mp_result2)                                                             \
 649                                                                                         \
 650   /* call doYouWantToGC() */                                                            \
 651   MAYBE_GC(R2_PTR & R4_PTR, name);                                                      \
 652                                                                                         \
 653   s1 = W_TO_INT(R1);                                                                    \
 654   d1 = R2;                                                                              \
 655   s2 = W_TO_INT(R3);                                                                    \
 656   d2 = R4;                                                                              \
 657                                                                                         \
 658   MP_INT__mp_alloc(mp_tmp1)     = W_TO_INT(StgArrWords_words(d1));                      \
 659   MP_INT__mp_size(mp_tmp1)      = (s1);                                                 \
 660   MP_INT__mp_d(mp_tmp1)         = BYTE_ARR_CTS(d1);                                     \
 661   MP_INT__mp_alloc(mp_tmp2)     = W_TO_INT(StgArrWords_words(d2));                      \
 662   MP_INT__mp_size(mp_tmp2)      = (s2);                                                 \
 663   MP_INT__mp_d(mp_tmp2)         = BYTE_ARR_CTS(d2);                                     \
 664                                                                                         \
 665   foreign "C" __gmpz_init(mp_result1 "ptr") [];                                               \
 666   foreign "C" __gmpz_init(mp_result2 "ptr") [];                                               \
 667                                                                                         \
 668   /* Perform the operation */                                                           \
 669   foreign "C" mp_fun(mp_result1 "ptr",mp_result2 "ptr",mp_tmp1 "ptr",mp_tmp2 "ptr") [];    \
 670                                                                                         \
 671   RET_NPNP(TO_W_(MP_INT__mp_size(mp_result1)),                                          \
 672            MP_INT__mp_d(mp_result1) - SIZEOF_StgArrWords,                               \
 673            TO_W_(MP_INT__mp_size(mp_result2)),                                          \
 674            MP_INT__mp_d(mp_result2) - SIZEOF_StgArrWords);                              \
 675 }
 676
 677 GMP_TAKE2_RET1(plusIntegerzh_fast,     __gmpz_add)
 678 GMP_TAKE2_RET1(minusIntegerzh_fast,    __gmpz_sub)
 679 GMP_TAKE2_RET1(timesIntegerzh_fast,    __gmpz_mul)
 680 GMP_TAKE2_RET1(gcdIntegerzh_fast,      __gmpz_gcd)
 681 GMP_TAKE2_RET1(quotIntegerzh_fast,     __gmpz_tdiv_q)
 682 GMP_TAKE2_RET1(remIntegerzh_fast,      __gmpz_tdiv_r)
 683 GMP_TAKE2_RET1(divExactIntegerzh_fast, __gmpz_divexact)
 684 GMP_TAKE2_RET1(andIntegerzh_fast,      __gmpz_and)
 685 GMP_TAKE2_RET1(orIntegerzh_fast,       __gmpz_ior)
 686 GMP_TAKE2_RET1(xorIntegerzh_fast,      __gmpz_xor)
 687 GMP_TAKE1_RET1(complementIntegerzh_fast, __gmpz_com)
 688
 689 GMP_TAKE2_RET2(quotRemIntegerzh_fast, __gmpz_tdiv_qr)
 690 GMP_TAKE2_RET2(divModIntegerzh_fast,  __gmpz_fdiv_qr)
 691
 692 #ifndef THREADED_RTS
 693 section "bss" {
 694   mp_tmp_w:  W_; // NB. mp_tmp_w is really an here mp_limb_t
 695 }
 696 #endif
 697
 698 gcdIntzh_fast
 699 {
 700     /* R1 = the first Int#; R2 = the second Int# */
 701     W_ r;
 702     FETCH_MP_TEMP(mp_tmp_w);
 703
 704     W_[mp_tmp_w] = R1;
 705     (r) = foreign "C" __gmpn_gcd_1(mp_tmp_w "ptr", 1, R2) [];
 706
 707     R1 = r;
 708     /* Result parked in R1, return via info-pointer at TOS */
 709     jump %ENTRY_CODE(Sp(0));
 710 }
 711
 712
 713 gcdIntegerIntzh_fast
 714 {
 715     /* R1 = s1; R2 = d1; R3 = the int */
 716     W_ s1;
 717     (s1) = foreign "C" __gmpn_gcd_1( BYTE_ARR_CTS(R2) "ptr", R1, R3) [];
 718     R1 = s1;
 719
 720     /* Result parked in R1, return via info-pointer at TOS */
 721     jump %ENTRY_CODE(Sp(0));
 722 }
 723
 724
 725 cmpIntegerIntzh_fast
 726 {
 727     /* R1 = s1; R2 = d1; R3 = the int */
 728     W_ usize, vsize, v_digit, u_digit;
 729
 730     usize = R1;
 731     vsize = 0;
 732     v_digit = R3;
 733
 734     // paraphrased from __gmpz_cmp_si() in the GMP sources
 735     if (%gt(v_digit,0)) {
 736         vsize = 1;
 737     } else {
 738         if (%lt(v_digit,0)) {
 739             vsize = -1;
 740             v_digit = -v_digit;
 741         }
 742     }
 743
 744     if (usize != vsize) {
 745         R1 = usize - vsize;
 746         jump %ENTRY_CODE(Sp(0));
 747     }
 748
 749     if (usize == 0) {
 750         R1 = 0;
 751         jump %ENTRY_CODE(Sp(0));
 752     }
 753
 754     u_digit = W_[BYTE_ARR_CTS(R2)];
 755
 756     if (u_digit == v_digit) {
 757         R1 = 0;
 758         jump %ENTRY_CODE(Sp(0));
 759     }
 760
 761     if (%gtu(u_digit,v_digit)) { // NB. unsigned: these are mp_limb_t's
 762         R1 = usize;
 763     } else {
 764         R1 = -usize;
 765     }
 766
 767     jump %ENTRY_CODE(Sp(0));
 768 }
 769
 770 cmpIntegerzh_fast
 771 {
 772     /* R1 = s1; R2 = d1; R3 = s2; R4 = d2 */
 773     W_ usize, vsize, size, up, vp;
 774     CInt cmp;
 775
 776     // paraphrased from __gmpz_cmp() in the GMP sources
 777     usize = R1;
 778     vsize = R3;
 779
 780     if (usize != vsize) {
 781         R1 = usize - vsize;
 782         jump %ENTRY_CODE(Sp(0));
 783     }
 784
 785     if (usize == 0) {
 786         R1 = 0;
 787         jump %ENTRY_CODE(Sp(0));
 788     }
 789
 790     if (%lt(usize,0)) { // NB. not <, which is unsigned
 791         size = -usize;
 792     } else {
 793         size = usize;
 794     }
 795
 796     up = BYTE_ARR_CTS(R2);
 797     vp = BYTE_ARR_CTS(R4);
 798
 799     (cmp) = foreign "C" __gmpn_cmp(up "ptr", vp "ptr", size) [];
 800
 801     if (cmp == 0 :: CInt) {
 802         R1 = 0;
 803         jump %ENTRY_CODE(Sp(0));
 804     }
 805
 806     if (%lt(cmp,0 :: CInt) == %lt(usize,0)) {
 807         R1 = 1;
 808     } else {
 809         R1 = (-1);
 810     }
 811     /* Result parked in R1, return via info-pointer at TOS */
 812     jump %ENTRY_CODE(Sp(0));
 813 }
 814
 815 integer2Intzh_fast
 816 {
 817     /* R1 = s; R2 = d */
 818     W_ r, s;
 819
 820     s = R1;
 821     if (s == 0) {
 822         r = 0;
 823     } else {
 824         r = W_[R2 + SIZEOF_StgArrWords];
 825         if (%lt(s,0)) {
 826             r = -r;
 827         }
 828     }
 829     /* Result parked in R1, return via info-pointer at TOS */
 830     R1 = r;
 831     jump %ENTRY_CODE(Sp(0));
 832 }
 833
 834 integer2Wordzh_fast
 835 {
 836   /* R1 = s; R2 = d */
 837   W_ r, s;
 838
 839   s = R1;
 840   if (s == 0) {
 841     r = 0;
 842   } else {
 843     r = W_[R2 + SIZEOF_StgArrWords];
 844     if (%lt(s,0)) {
 845         r = -r;
 846     }
 847   }
 848   /* Result parked in R1, return via info-pointer at TOS */
 849   R1 = r;
 850   jump %ENTRY_CODE(Sp(0));
 851 }
 852
 853 decodeFloatzh_fast
 854 {
 855     W_ p;
 856     F_ arg;
 857     FETCH_MP_TEMP(mp_tmp1);
 858     FETCH_MP_TEMP(mp_tmp_w);
 859
 860     /* arguments: F1 = Float# */
 861     arg = F1;
 862
 863     ALLOC_PRIM( SIZEOF_StgArrWords + WDS(1), NO_PTRS, decodeFloatzh_fast );
 864
 865     /* Be prepared to tell Lennart-coded __decodeFloat
 866        where mantissa._mp_d can be put (it does not care about the rest) */
 867     p = Hp - SIZEOF_StgArrWords;
 868     SET_HDR(p,stg_ARR_WORDS_info,W_[CCCS]);
 869     StgArrWords_words(p) = 1;
 870     MP_INT__mp_d(mp_tmp1) = BYTE_ARR_CTS(p);
 871
 872     /* Perform the operation */
 873     foreign "C" __decodeFloat(mp_tmp1 "ptr",mp_tmp_w "ptr" ,arg) [];
 874
 875     /* returns: (Int# (expn), Int#, ByteArray#) */
 876     RET_NNP(W_[mp_tmp_w], TO_W_(MP_INT__mp_size(mp_tmp1)), p);
 877 }
 878
 879 decodeFloatzuIntzh_fast
 880 {
 881     W_ p;
 882     F_ arg;
 883     FETCH_MP_TEMP(mp_tmp1);
 884     FETCH_MP_TEMP(mp_tmp_w);
 885
 886     /* arguments: F1 = Float# */
 887     arg = F1;
 888
 889     /* Perform the operation */
 890     foreign "C" __decodeFloat_Int(mp_tmp1 "ptr", mp_tmp_w "ptr", arg) [];
 891
 892     /* returns: (Int# (mantissa), Int# (exponent)) */
 893     RET_NN(W_[mp_tmp1], W_[mp_tmp_w]);
 894 }
 895
 896 #define DOUBLE_MANTISSA_SIZE SIZEOF_DOUBLE
 897 #define ARR_SIZE (SIZEOF_StgArrWords + DOUBLE_MANTISSA_SIZE)
 898
 899 decodeDoublezh_fast
 900 {
 901     D_ arg;
 902     W_ p;
 903     FETCH_MP_TEMP(mp_tmp1);
 904     FETCH_MP_TEMP(mp_tmp_w);
 905
 906     /* arguments: D1 = Double# */
 907     arg = D1;
 908
 909     ALLOC_PRIM( ARR_SIZE, NO_PTRS, decodeDoublezh_fast );
 910
 911     /* Be prepared to tell Lennart-coded __decodeDouble
 912        where mantissa.d can be put (it does not care about the rest) */
 913     p = Hp - ARR_SIZE + WDS(1);
 914     SET_HDR(p, stg_ARR_WORDS_info, W_[CCCS]);
 915     StgArrWords_words(p) = BYTES_TO_WDS(DOUBLE_MANTISSA_SIZE);
 916     MP_INT__mp_d(mp_tmp1) = BYTE_ARR_CTS(p);
 917
 918     /* Perform the operation */
 919     foreign "C" __decodeDouble(mp_tmp1 "ptr", mp_tmp_w "ptr",arg) [];
 920
 921     /* returns: (Int# (expn), Int#, ByteArray#) */
 922     RET_NNP(W_[mp_tmp_w], TO_W_(MP_INT__mp_size(mp_tmp1)), p);
 923 }
 924
 925 decodeDoublezu2Intzh_fast
 926 {
 927     D_ arg;
 928     W_ p;
 929     FETCH_MP_TEMP(mp_tmp1);
 930     FETCH_MP_TEMP(mp_tmp2);
 931     FETCH_MP_TEMP(mp_result1);
 932     FETCH_MP_TEMP(mp_result2);
 933
 934     /* arguments: D1 = Double# */
 935     arg = D1;
 936
 937     /* Perform the operation */
 938     foreign "C" __decodeDouble_2Int(mp_tmp1 "ptr", mp_tmp2 "ptr",
 939                                     mp_result1 "ptr", mp_result2 "ptr",
 940                                     arg) [];
 941
 942     /* returns:
 943        (Int# (mant sign), Word# (mant high), Word# (mant low), Int# (expn)) */
 944     RET_NNNN(W_[mp_tmp1], W_[mp_tmp2], W_[mp_result1], W_[mp_result2]);
 945 }
 946
 947 /* -----------------------------------------------------------------------------
 948  * Concurrency primitives
 949  * -------------------------------------------------------------------------- */
 950
 951 forkzh_fast
 952 {
 953   /* args: R1 = closure to spark */
 954
 955   MAYBE_GC(R1_PTR, forkzh_fast);
 956
 957   W_ closure;
 958   W_ threadid;
 959   closure = R1;
 960
 961   ("ptr" threadid) = foreign "C" createIOThread( MyCapability() "ptr",
 962                                 RtsFlags_GcFlags_initialStkSize(RtsFlags),
 963                                 closure "ptr") [];
 964
 965   /* start blocked if the current thread is blocked */
 966   StgTSO_flags(threadid) =
 967      StgTSO_flags(threadid) |  (StgTSO_flags(CurrentTSO) &
 968                                 (TSO_BLOCKEX::I32 | TSO_INTERRUPTIBLE::I32));
 969
 970   foreign "C" scheduleThread(MyCapability() "ptr", threadid "ptr") [];
 971
 972   // switch at the earliest opportunity
 973   CInt[context_switch] = 1 :: CInt;
 974
 975   RET_P(threadid);
 976 }
 977
 978 forkOnzh_fast
 979 {
 980   /* args: R1 = cpu, R2 = closure to spark */
 981
 982   MAYBE_GC(R2_PTR, forkOnzh_fast);
 983
 984   W_ cpu;
 985   W_ closure;
 986   W_ threadid;
 987   cpu = R1;
 988   closure = R2;
 989
 990   ("ptr" threadid) = foreign "C" createIOThread( MyCapability() "ptr",
 991                                 RtsFlags_GcFlags_initialStkSize(RtsFlags),
 992                                 closure "ptr") [];
 993
 994   /* start blocked if the current thread is blocked */
 995   StgTSO_flags(threadid) =
 996      StgTSO_flags(threadid) |  (StgTSO_flags(CurrentTSO) &
 997                                 (TSO_BLOCKEX::I32 | TSO_INTERRUPTIBLE::I32));
 998
 999   foreign "C" scheduleThreadOn(MyCapability() "ptr", cpu, threadid "ptr") [];
1000
1001   // switch at the earliest opportunity
1002   CInt[context_switch] = 1 :: CInt;
1003
1004   RET_P(threadid);
1005 }
1006
1007 yieldzh_fast
1008 {
1009   jump stg_yield_noregs;
1010 }
1011
1012 myThreadIdzh_fast
1013 {
1014   /* no args. */
1015   RET_P(CurrentTSO);
1016 }
1017
1018 labelThreadzh_fast
1019 {
1020   /* args:
1021         R1 = ThreadId#
1022         R2 = Addr# */
1023 #ifdef DEBUG
1024   foreign "C" labelThread(R1 "ptr", R2 "ptr") [];
1025 #endif
1026   jump %ENTRY_CODE(Sp(0));
1027 }
1028
1029 isCurrentThreadBoundzh_fast
1030 {
1031   /* no args */
1032   W_ r;
1033   (r) = foreign "C" isThreadBound(CurrentTSO) [];
1034   RET_N(r);
1035 }
1036
1037
1038 /* -----------------------------------------------------------------------------
1039  * TVar primitives
1040  * -------------------------------------------------------------------------- */
1041
1042 #ifdef REG_R1
1043 #define SP_OFF 0
1044 #define IF_NOT_REG_R1(x)
1045 #else
1046 #define SP_OFF 1
1047 #define IF_NOT_REG_R1(x) x
1048 #endif
1049
1050 // Catch retry frame ------------------------------------------------------------
1051
1052 INFO_TABLE_RET(stg_catch_retry_frame, CATCH_RETRY_FRAME,
1053 #if defined(PROFILING)
1054   W_ unused1, W_ unused2,
1055 #endif
1056   W_ unused3, "ptr" W_ unused4, "ptr" W_ unused5)
1057 {
1058    W_ r, frame, trec, outer;
1059    IF_NOT_REG_R1(W_ rval;  rval = Sp(0);  Sp_adj(1); )
1060
1061    frame = Sp;
1062    trec = StgTSO_trec(CurrentTSO);
1063    ("ptr" outer) = foreign "C" stmGetEnclosingTRec(trec "ptr") [];
1064    (r) = foreign "C" stmCommitNestedTransaction(MyCapability() "ptr", trec "ptr") [];
1065    if (r != 0) {
1066      /* Succeeded (either first branch or second branch) */
1067      StgTSO_trec(CurrentTSO) = outer;
1068      Sp = Sp + SIZEOF_StgCatchRetryFrame;
1069      IF_NOT_REG_R1(Sp_adj(-1); Sp(0) = rval;)
1070      jump %ENTRY_CODE(Sp(SP_OFF));
1071    } else {
1072      /* Did not commit: re-execute */
1073      W_ new_trec;
1074      ("ptr" new_trec) = foreign "C" stmStartTransaction(MyCapability() "ptr", outer "ptr") [];
1075      StgTSO_trec(CurrentTSO) = new_trec;
1076      if (StgCatchRetryFrame_running_alt_code(frame) != 0::I32) {
1077        R1 = StgCatchRetryFrame_alt_code(frame);
1078      } else {
1079        R1 = StgCatchRetryFrame_first_code(frame);
1080      }
1081      jump stg_ap_v_fast;
1082    }
1083 }
1084
1085
1086 // Atomically frame ------------------------------------------------------------
1087
1088 INFO_TABLE_RET(stg_atomically_frame, ATOMICALLY_FRAME,
1089 #if defined(PROFILING)
1090   W_ unused1, W_ unused2,
1091 #endif
1092   "ptr" W_ unused3, "ptr" W_ unused4)
1093 {
1094   W_ frame, trec, valid, next_invariant, q, outer;
1095   IF_NOT_REG_R1(W_ rval;  rval = Sp(0);  Sp_adj(1); )
1096
1097   frame = Sp;
1098   trec = StgTSO_trec(CurrentTSO);
1099   ("ptr" outer) = foreign "C" stmGetEnclosingTRec(trec "ptr") [];
1100
1101   if (outer == NO_TREC) {
1102     /* First time back at the atomically frame -- pick up invariants */
1103     ("ptr" q) = foreign "C" stmGetInvariantsToCheck(MyCapability() "ptr", trec "ptr") [];
1104     StgAtomicallyFrame_next_invariant_to_check(frame) = q;
1105
1106   } else {
1107     /* Second/subsequent time back at the atomically frame -- abort the
1108      * tx that's checking the invariant and move on to the next one */
1109     StgTSO_trec(CurrentTSO) = outer;
1110     q = StgAtomicallyFrame_next_invariant_to_check(frame);
1111     StgInvariantCheckQueue_my_execution(q) = trec;
1112     foreign "C" stmAbortTransaction(MyCapability() "ptr", trec "ptr") [];
1113     /* Don't free trec -- it's linked from q and will be stashed in the
1114      * invariant if we eventually commit. */
1115     q = StgInvariantCheckQueue_next_queue_entry(q);
1116     StgAtomicallyFrame_next_invariant_to_check(frame) = q;
1117     trec = outer;
1118   }
1119
1120   q = StgAtomicallyFrame_next_invariant_to_check(frame);
1121
1122   if (q != END_INVARIANT_CHECK_QUEUE) {
1123     /* We can't commit yet: another invariant to check */
1124     ("ptr" trec) = foreign "C" stmStartTransaction(MyCapability() "ptr", trec "ptr") [];
1125     StgTSO_trec(CurrentTSO) = trec;
1126
1127     next_invariant = StgInvariantCheckQueue_invariant(q);
1128     R1 = StgAtomicInvariant_code(next_invariant);
1129     jump stg_ap_v_fast;
1130
1131   } else {
1132
1133     /* We've got no more invariants to check, try to commit */
1134     (valid) = foreign "C" stmCommitTransaction(MyCapability() "ptr", trec "ptr") [];
1135     if (valid != 0) {
1136       /* Transaction was valid: commit succeeded */
1137       StgTSO_trec(CurrentTSO) = NO_TREC;
1138       Sp = Sp + SIZEOF_StgAtomicallyFrame;
1139       IF_NOT_REG_R1(Sp_adj(-1); Sp(0) = rval;)
1140       jump %ENTRY_CODE(Sp(SP_OFF));
1141     } else {
1142       /* Transaction was not valid: try again */
1143       ("ptr" trec) = foreign "C" stmStartTransaction(MyCapability() "ptr", NO_TREC "ptr") [];
1144       StgTSO_trec(CurrentTSO) = trec;
1145       StgAtomicallyFrame_next_invariant_to_check(frame) = END_INVARIANT_CHECK_QUEUE;
1146       R1 = StgAtomicallyFrame_code(frame);
1147       jump stg_ap_v_fast;
1148     }
1149   }
1150 }
1151
1152 INFO_TABLE_RET(stg_atomically_waiting_frame, ATOMICALLY_FRAME,
1153 #if defined(PROFILING)
1154   W_ unused1, W_ unused2,
1155 #endif
1156   "ptr" W_ unused3, "ptr" W_ unused4)
1157 {
1158   W_ frame, trec, valid;
1159   IF_NOT_REG_R1(W_ rval;  rval = Sp(0);  Sp_adj(1); )
1160
1161   frame = Sp;
1162
1163   /* The TSO is currently waiting: should we stop waiting? */
1164   (valid) = foreign "C" stmReWait(MyCapability() "ptr", CurrentTSO "ptr") [];
1165   if (valid != 0) {
1166     /* Previous attempt is still valid: no point trying again yet */
1167           IF_NOT_REG_R1(Sp_adj(-2);
1168                         Sp(1) = stg_NO_FINALIZER_closure;
1169                         Sp(0) = stg_ut_1_0_unreg_info;)
1170     jump stg_block_noregs;
1171   } else {
1172     /* Previous attempt is no longer valid: try again */
1173     ("ptr" trec) = foreign "C" stmStartTransaction(MyCapability() "ptr", NO_TREC "ptr") [];
1174     StgTSO_trec(CurrentTSO) = trec;
1175     StgHeader_info(frame) = stg_atomically_frame_info;
1176     R1 = StgAtomicallyFrame_code(frame);
1177     jump stg_ap_v_fast;
1178   }
1179 }
1180
1181 // STM catch frame --------------------------------------------------------------
1182
1183 #ifdef REG_R1
1184 #define SP_OFF 0
1185 #else
1186 #define SP_OFF 1
1187 #endif
1188
1189 /* Catch frames are very similar to update frames, but when entering
1190  * one we just pop the frame off the stack and perform the correct
1191  * kind of return to the activation record underneath us on the stack.
1192  */
1193
1194 INFO_TABLE_RET(stg_catch_stm_frame, CATCH_STM_FRAME,
1195 #if defined(PROFILING)
1196   W_ unused1, W_ unused2,
1197 #endif
1198   "ptr" W_ unused3, "ptr" W_ unused4)
1199    {
1200       IF_NOT_REG_R1(W_ rval;  rval = Sp(0);  Sp_adj(1); )
1201       W_ r, frame, trec, outer;
1202       frame = Sp;
1203       trec = StgTSO_trec(CurrentTSO);
1204       ("ptr" outer) = foreign "C" stmGetEnclosingTRec(trec "ptr") [];
1205       (r) = foreign "C" stmCommitNestedTransaction(MyCapability() "ptr", trec "ptr") [];
1206       if (r != 0) {
1207         /* Commit succeeded */
1208         StgTSO_trec(CurrentTSO) = outer;
1209         Sp = Sp + SIZEOF_StgCatchSTMFrame;
1210         IF_NOT_REG_R1(Sp_adj(-1); Sp(0) = rval;)
1211         jump Sp(SP_OFF);
1212       } else {
1213         /* Commit failed */
1214         W_ new_trec;
1215         ("ptr" new_trec) = foreign "C" stmStartTransaction(MyCapability() "ptr", outer "ptr") [];
1216         StgTSO_trec(CurrentTSO) = new_trec;
1217         R1 = StgCatchSTMFrame_code(frame);
1218         jump stg_ap_v_fast;
1219       }
1220    }
1221
1222
1223 // Primop definition ------------------------------------------------------------
1224
1225 atomicallyzh_fast
1226 {
1227   W_ frame;
1228   W_ old_trec;
1229   W_ new_trec;
1230
1231   // stmStartTransaction may allocate
1232   MAYBE_GC (R1_PTR, atomicallyzh_fast);
1233
1234   /* Args: R1 = m :: STM a */
1235   STK_CHK_GEN(SIZEOF_StgAtomicallyFrame + WDS(1), R1_PTR, atomicallyzh_fast);
1236
1237   old_trec = StgTSO_trec(CurrentTSO);
1238
1239   /* Nested transactions are not allowed; raise an exception */
1240   if (old_trec != NO_TREC) {
1241      R1 = base_GHCziIOBase_NestedAtomically_closure;
1242      jump raisezh_fast;
1243   }
1244
1245   /* Set up the atomically frame */
1246   Sp = Sp - SIZEOF_StgAtomicallyFrame;
1247   frame = Sp;
1248
1249   SET_HDR(frame,stg_atomically_frame_info, W_[CCCS]);
1250   StgAtomicallyFrame_code(frame) = R1;
1251   StgAtomicallyFrame_next_invariant_to_check(frame) = END_INVARIANT_CHECK_QUEUE;
1252
1253   /* Start the memory transcation */
1254   ("ptr" new_trec) = foreign "C" stmStartTransaction(MyCapability() "ptr", old_trec "ptr") [R1];
1255   StgTSO_trec(CurrentTSO) = new_trec;
1256
1257   /* Apply R1 to the realworld token */
1258   jump stg_ap_v_fast;
1259 }
1260
1261
1262 catchSTMzh_fast
1263 {
1264   W_ frame;
1265
1266   /* Args: R1 :: STM a */
1267   /* Args: R2 :: Exception -> STM a */
1268   STK_CHK_GEN(SIZEOF_StgCatchSTMFrame + WDS(1), R1_PTR & R2_PTR, catchSTMzh_fast);
1269
1270   /* Set up the catch frame */
1271   Sp = Sp - SIZEOF_StgCatchSTMFrame;
1272   frame = Sp;
1273
1274   SET_HDR(frame, stg_catch_stm_frame_info, W_[CCCS]);
1275   StgCatchSTMFrame_handler(frame) = R2;
1276   StgCatchSTMFrame_code(frame) = R1;
1277
1278   /* Start a nested transaction to run the body of the try block in */
1279   W_ cur_trec;
1280   W_ new_trec;
1281   cur_trec = StgTSO_trec(CurrentTSO);
1282   ("ptr" new_trec) = foreign "C" stmStartTransaction(MyCapability() "ptr", cur_trec "ptr");
1283   StgTSO_trec(CurrentTSO) = new_trec;
1284
1285   /* Apply R1 to the realworld token */
1286   jump stg_ap_v_fast;
1287 }
1288
1289
1290 catchRetryzh_fast
1291 {
1292   W_ frame;
1293   W_ new_trec;
1294   W_ trec;
1295
1296   // stmStartTransaction may allocate
1297   MAYBE_GC (R1_PTR & R2_PTR, catchRetryzh_fast);
1298
1299   /* Args: R1 :: STM a */
1300   /* Args: R2 :: STM a */
1301   STK_CHK_GEN(SIZEOF_StgCatchRetryFrame + WDS(1), R1_PTR & R2_PTR, catchRetryzh_fast);
1302
1303   /* Start a nested transaction within which to run the first code */
1304   trec = StgTSO_trec(CurrentTSO);
1305   ("ptr" new_trec) = foreign "C" stmStartTransaction(MyCapability() "ptr", trec "ptr") [R1,R2];
1306   StgTSO_trec(CurrentTSO) = new_trec;
1307
1308   /* Set up the catch-retry frame */
1309   Sp = Sp - SIZEOF_StgCatchRetryFrame;
1310   frame = Sp;
1311
1312   SET_HDR(frame, stg_catch_retry_frame_info, W_[CCCS]);
1313   StgCatchRetryFrame_running_alt_code(frame) = 0 :: CInt; // false;
1314   StgCatchRetryFrame_first_code(frame) = R1;
1315   StgCatchRetryFrame_alt_code(frame) = R2;
1316
1317   /* Apply R1 to the realworld token */
1318   jump stg_ap_v_fast;
1319 }
1320
1321
1322 retryzh_fast
1323 {
1324   W_ frame_type;
1325   W_ frame;
1326   W_ trec;
1327   W_ outer;
1328   W_ r;
1329
1330   MAYBE_GC (NO_PTRS, retryzh_fast); // STM operations may allocate
1331
1332   // Find the enclosing ATOMICALLY_FRAME or CATCH_RETRY_FRAME
1333 retry_pop_stack:
1334   StgTSO_sp(CurrentTSO) = Sp;
1335   (frame_type) = foreign "C" findRetryFrameHelper(CurrentTSO "ptr") [];
1336   Sp = StgTSO_sp(CurrentTSO);
1337   frame = Sp;
1338   trec = StgTSO_trec(CurrentTSO);
1339   ("ptr" outer) = foreign "C" stmGetEnclosingTRec(trec "ptr") [];
1340
1341   if (frame_type == CATCH_RETRY_FRAME) {
1342     // The retry reaches a CATCH_RETRY_FRAME before the atomic frame
1343     ASSERT(outer != NO_TREC);
1344     // Abort the transaction attempting the current branch
1345     foreign "C" stmAbortTransaction(MyCapability() "ptr", trec "ptr") [];
1346     foreign "C" stmFreeAbortedTRec(MyCapability() "ptr", trec "ptr") [];
1347     if (!StgCatchRetryFrame_running_alt_code(frame) != 0::I32) {
1348       // Retry in the first branch: try the alternative
1349       ("ptr" trec) = foreign "C" stmStartTransaction(MyCapability() "ptr", outer "ptr") [];
1350       StgTSO_trec(CurrentTSO) = trec;
1351       StgCatchRetryFrame_running_alt_code(frame) = 1 :: CInt; // true;
1352       R1 = StgCatchRetryFrame_alt_code(frame);
1353       jump stg_ap_v_fast;
1354     } else {
1355       // Retry in the alternative code: propagate the retry
1356       StgTSO_trec(CurrentTSO) = outer;
1357       Sp = Sp + SIZEOF_StgCatchRetryFrame;
1358       goto retry_pop_stack;
1359     }
1360   }
1361
1362   // We've reached the ATOMICALLY_FRAME: attempt to wait
1363   ASSERT(frame_type == ATOMICALLY_FRAME);
1364   if (outer != NO_TREC) {
1365     // We called retry while checking invariants, so abort the current
1366     // invariant check (merging its TVar accesses into the parents read
1367     // set so we'll wait on them)
1368     foreign "C" stmAbortTransaction(MyCapability() "ptr", trec "ptr") [];
1369     foreign "C" stmFreeAbortedTRec(MyCapability() "ptr", trec "ptr") [];
1370     trec = outer;
1371     StgTSO_trec(CurrentTSO) = trec;
1372     ("ptr" outer) = foreign "C" stmGetEnclosingTRec(trec "ptr") [];
1373   }
1374   ASSERT(outer == NO_TREC);
1375
1376   (r) = foreign "C" stmWait(MyCapability() "ptr", CurrentTSO "ptr", trec "ptr") [];
1377   if (r != 0) {
1378     // Transaction was valid: stmWait put us on the TVars' queues, we now block
1379     StgHeader_info(frame) = stg_atomically_waiting_frame_info;
1380     Sp = frame;
1381     // Fix up the stack in the unregisterised case: the return convention is different.
1382     IF_NOT_REG_R1(Sp_adj(-2);
1383                   Sp(1) = stg_NO_FINALIZER_closure;
1384                   Sp(0) = stg_ut_1_0_unreg_info;)
1385     R3 = trec; // passing to stmWaitUnblock()
1386     jump stg_block_stmwait;
1387   } else {
1388     // Transaction was not valid: retry immediately
1389     ("ptr" trec) = foreign "C" stmStartTransaction(MyCapability() "ptr", outer "ptr") [];
1390     StgTSO_trec(CurrentTSO) = trec;
1391     R1 = StgAtomicallyFrame_code(frame);
1392     Sp = frame;
1393     jump stg_ap_v_fast;
1394   }
1395 }
1396
1397
1398 checkzh_fast
1399 {
1400   W_ trec, closure;
1401
1402   /* Args: R1 = invariant closure */
1403   MAYBE_GC (R1_PTR, checkzh_fast);
1404
1405   trec = StgTSO_trec(CurrentTSO);
1406   closure = R1;
1407   foreign "C" stmAddInvariantToCheck(MyCapability() "ptr",
1408                                      trec "ptr",
1409                                      closure "ptr") [];
1410
1411   jump %ENTRY_CODE(Sp(0));
1412 }
1413
1414
1415 newTVarzh_fast
1416 {
1417   W_ tv;
1418   W_ new_value;
1419
1420   /* Args: R1 = initialisation value */
1421
1422   MAYBE_GC (R1_PTR, newTVarzh_fast);
1423   new_value = R1;
1424   ("ptr" tv) = foreign "C" stmNewTVar(MyCapability() "ptr", new_value "ptr") [];
1425   RET_P(tv);
1426 }
1427
1428
1429 readTVarzh_fast
1430 {
1431   W_ trec;
1432   W_ tvar;
1433   W_ result;
1434
1435   /* Args: R1 = TVar closure */
1436
1437   MAYBE_GC (R1_PTR, readTVarzh_fast); // Call to stmReadTVar may allocate
1438   trec = StgTSO_trec(CurrentTSO);
1439   tvar = R1;
1440   ("ptr" result) = foreign "C" stmReadTVar(MyCapability() "ptr", trec "ptr", tvar "ptr") [];
1441
1442   RET_P(result);
1443 }
1444
1445
1446 writeTVarzh_fast
1447 {
1448   W_ trec;
1449   W_ tvar;
1450   W_ new_value;
1451
1452   /* Args: R1 = TVar closure */
1453   /*       R2 = New value    */
1454
1455   MAYBE_GC (R1_PTR & R2_PTR, writeTVarzh_fast); // Call to stmWriteTVar may allocate
1456   trec = StgTSO_trec(CurrentTSO);
1457   tvar = R1;
1458   new_value = R2;
1459   foreign "C" stmWriteTVar(MyCapability() "ptr", trec "ptr", tvar "ptr", new_value "ptr") [];
1460
1461   jump %ENTRY_CODE(Sp(0));
1462 }
1463
1464
1465 /* -----------------------------------------------------------------------------
1466  * MVar primitives
1467  *
1468  * take & putMVar work as follows.  Firstly, an important invariant:
1469  *
1470  *    If the MVar is full, then the blocking queue contains only
1471  *    threads blocked on putMVar, and if the MVar is empty then the
1472  *    blocking queue contains only threads blocked on takeMVar.
1473  *
1474  * takeMvar:
1475  *    MVar empty : then add ourselves to the blocking queue
1476  *    MVar full  : remove the value from the MVar, and
1477  *                 blocking queue empty     : return
1478  *                 blocking queue non-empty : perform the first blocked putMVar
1479  *                                            from the queue, and wake up the
1480  *                                            thread (MVar is now full again)
1481  *
1482  * putMVar is just the dual of the above algorithm.
1483  *
1484  * How do we "perform a putMVar"?  Well, we have to fiddle around with
1485  * the stack of the thread waiting to do the putMVar.  See
1486  * stg_block_putmvar and stg_block_takemvar in HeapStackCheck.c for
1487  * the stack layout, and the PerformPut and PerformTake macros below.
1488  *
1489  * It is important that a blocked take or put is woken up with the
1490  * take/put already performed, because otherwise there would be a
1491  * small window of vulnerability where the thread could receive an
1492  * exception and never perform its take or put, and we'd end up with a
1493  * deadlock.
1494  *
1495  * -------------------------------------------------------------------------- */
1496
1497 isEmptyMVarzh_fast
1498 {
1499     /* args: R1 = MVar closure */
1500
1501     if (StgMVar_value(R1) == stg_END_TSO_QUEUE_closure) {
1502         RET_N(1);
1503     } else {
1504         RET_N(0);
1505     }
1506 }
1507
1508 newMVarzh_fast
1509 {
1510     /* args: none */
1511     W_ mvar;
1512
1513     ALLOC_PRIM ( SIZEOF_StgMVar, NO_PTRS, newMVarzh_fast );
1514
1515     mvar = Hp - SIZEOF_StgMVar + WDS(1);
1516     SET_HDR(mvar,stg_MVAR_DIRTY_info,W_[CCCS]);
1517         // MVARs start dirty: generation 0 has no mutable list
1518     StgMVar_head(mvar)  = stg_END_TSO_QUEUE_closure;
1519     StgMVar_tail(mvar)  = stg_END_TSO_QUEUE_closure;
1520     StgMVar_value(mvar) = stg_END_TSO_QUEUE_closure;
1521     RET_P(mvar);
1522 }
1523
1524
1525 /* If R1 isn't available, pass it on the stack */
1526 #ifdef REG_R1
1527 #define PerformTake(tso, value)                         \
1528     W_[StgTSO_sp(tso) + WDS(1)] = value;                \
1529     W_[StgTSO_sp(tso) + WDS(0)] = stg_gc_unpt_r1_info;
1530 #else
1531 #define PerformTake(tso, value)                                 \
1532     W_[StgTSO_sp(tso) + WDS(1)] = value;                        \
1533     W_[StgTSO_sp(tso) + WDS(0)] = stg_ut_1_0_unreg_info;
1534 #endif
1535
1536 #define PerformPut(tso,lval)                    \
1537     StgTSO_sp(tso) = StgTSO_sp(tso) + WDS(3);   \
1538     lval = W_[StgTSO_sp(tso) - WDS(1)];
1539
1540 takeMVarzh_fast
1541 {
1542     W_ mvar, val, info, tso;
1543
1544     /* args: R1 = MVar closure */
1545     mvar = R1;
1546
1547 #if defined(THREADED_RTS)
1548     ("ptr" info) = foreign "C" lockClosure(mvar "ptr") [];
1549 #else
1550     info = GET_INFO(mvar);
1551 #endif
1552
1553     if (info == stg_MVAR_CLEAN_info) {
1554         foreign "C" dirty_MVAR(BaseReg "ptr", mvar "ptr");
1555     }
1556
1557     /* If the MVar is empty, put ourselves on its blocking queue,
1558      * and wait until we're woken up.
1559      */
1560     if (StgMVar_value(mvar) == stg_END_TSO_QUEUE_closure) {
1561         if (StgMVar_head(mvar) == stg_END_TSO_QUEUE_closure) {
1562             StgMVar_head(mvar) = CurrentTSO;
1563         } else {
1564             foreign "C" setTSOLink(MyCapability() "ptr", StgMVar_tail(mvar),
1565                                    CurrentTSO);
1566         }
1567         StgTSO__link(CurrentTSO)       = stg_END_TSO_QUEUE_closure;
1568         StgTSO_why_blocked(CurrentTSO) = BlockedOnMVar::I16;
1569         StgTSO_block_info(CurrentTSO)  = mvar;
1570         StgMVar_tail(mvar) = CurrentTSO;
1571
1572         jump stg_block_takemvar;
1573   }
1574
1575   /* we got the value... */
1576   val = StgMVar_value(mvar);
1577
1578   if (StgMVar_head(mvar) != stg_END_TSO_QUEUE_closure)
1579   {
1580       /* There are putMVar(s) waiting...
1581        * wake up the first thread on the queue
1582        */
1583       ASSERT(StgTSO_why_blocked(StgMVar_head(mvar)) == BlockedOnMVar::I16);
1584
1585       /* actually perform the putMVar for the thread that we just woke up */
1586       tso = StgMVar_head(mvar);
1587       PerformPut(tso,StgMVar_value(mvar));
1588
1589       if (StgTSO_flags(tso) & TSO_DIRTY == 0) {
1590           foreign "C" dirty_TSO(MyCapability(), tso);
1591       }
1592
1593       ("ptr" tso) = foreign "C" unblockOne_(MyCapability() "ptr",
1594                                             StgMVar_head(mvar) "ptr", 1) [];
1595       StgMVar_head(mvar) = tso;
1596
1597       if (StgMVar_head(mvar) == stg_END_TSO_QUEUE_closure) {
1598           StgMVar_tail(mvar) = stg_END_TSO_QUEUE_closure;
1599       }
1600
1601 #if defined(THREADED_RTS)
1602       unlockClosure(mvar, stg_MVAR_DIRTY_info);
1603 #else
1604       SET_INFO(mvar,stg_MVAR_DIRTY_info);
1605 #endif
1606       RET_P(val);
1607   }
1608   else
1609   {
1610       /* No further putMVars, MVar is now empty */
1611       StgMVar_value(mvar) = stg_END_TSO_QUEUE_closure;
1612
1613 #if defined(THREADED_RTS)
1614       unlockClosure(mvar, stg_MVAR_DIRTY_info);
1615 #else
1616       SET_INFO(mvar,stg_MVAR_DIRTY_info);
1617 #endif
1618
1619       RET_P(val);
1620   }
1621 }
1622
1623
1624 tryTakeMVarzh_fast
1625 {
1626     W_ mvar, val, info, tso;
1627
1628     /* args: R1 = MVar closure */
1629
1630     mvar = R1;
1631
1632 #if defined(THREADED_RTS)
1633     ("ptr" info) = foreign "C" lockClosure(mvar "ptr") [];
1634 #else
1635     info = GET_INFO(mvar);
1636 #endif
1637
1638     if (StgMVar_value(mvar) == stg_END_TSO_QUEUE_closure) {
1639 #if defined(THREADED_RTS)
1640         unlockClosure(mvar, info);
1641 #endif
1642         /* HACK: we need a pointer to pass back,
1643          * so we abuse NO_FINALIZER_closure
1644          */
1645         RET_NP(0, stg_NO_FINALIZER_closure);
1646     }
1647
1648     if (info == stg_MVAR_CLEAN_info) {
1649         foreign "C" dirty_MVAR(BaseReg "ptr", mvar "ptr");
1650     }
1651
1652     /* we got the value... */
1653     val = StgMVar_value(mvar);
1654
1655     if (StgMVar_head(mvar) != stg_END_TSO_QUEUE_closure) {
1656
1657         /* There are putMVar(s) waiting...
1658          * wake up the first thread on the queue
1659          */
1660         ASSERT(StgTSO_why_blocked(StgMVar_head(mvar)) == BlockedOnMVar::I16);
1661
1662         /* actually perform the putMVar for the thread that we just woke up */
1663         tso = StgMVar_head(mvar);
1664         PerformPut(tso,StgMVar_value(mvar));
1665         if (StgTSO_flags(tso) & TSO_DIRTY == 0) {
1666             foreign "C" dirty_TSO(MyCapability(), tso);
1667         }
1668
1669         ("ptr" tso) = foreign "C" unblockOne_(MyCapability() "ptr",
1670                                               StgMVar_head(mvar) "ptr", 1) [];
1671         StgMVar_head(mvar) = tso;
1672
1673         if (StgMVar_head(mvar) == stg_END_TSO_QUEUE_closure) {
1674             StgMVar_tail(mvar) = stg_END_TSO_QUEUE_closure;
1675         }
1676 #if defined(THREADED_RTS)
1677         unlockClosure(mvar, stg_MVAR_DIRTY_info);
1678 #else
1679         SET_INFO(mvar,stg_MVAR_DIRTY_info);
1680 #endif
1681     }
1682     else
1683     {
1684         /* No further putMVars, MVar is now empty */
1685         StgMVar_value(mvar) = stg_END_TSO_QUEUE_closure;
1686 #if defined(THREADED_RTS)
1687         unlockClosure(mvar, stg_MVAR_DIRTY_info);
1688 #else
1689         SET_INFO(mvar,stg_MVAR_DIRTY_info);
1690 #endif
1691     }
1692
1693     RET_NP(1, val);
1694 }
1695
1696
1697 putMVarzh_fast
1698 {
1699     W_ mvar, info, tso;
1700
1701     /* args: R1 = MVar, R2 = value */
1702     mvar = R1;
1703
1704 #if defined(THREADED_RTS)
1705     ("ptr" info) = foreign "C" lockClosure(mvar "ptr") [R2];
1706 #else
1707     info = GET_INFO(mvar);
1708 #endif
1709
1710     if (info == stg_MVAR_CLEAN_info) {
1711         foreign "C" dirty_MVAR(BaseReg "ptr", mvar "ptr");
1712     }
1713
1714     if (StgMVar_value(mvar) != stg_END_TSO_QUEUE_closure) {
1715         if (StgMVar_head(mvar) == stg_END_TSO_QUEUE_closure) {
1716             StgMVar_head(mvar) = CurrentTSO;
1717         } else {
1718             foreign "C" setTSOLink(MyCapability() "ptr", StgMVar_tail(mvar),
1719                                    CurrentTSO);
1720         }
1721         StgTSO__link(CurrentTSO)       = stg_END_TSO_QUEUE_closure;
1722         StgTSO_why_blocked(CurrentTSO) = BlockedOnMVar::I16;
1723         StgTSO_block_info(CurrentTSO)  = mvar;
1724         StgMVar_tail(mvar) = CurrentTSO;
1725
1726         jump stg_block_putmvar;
1727     }
1728
1729     if (StgMVar_head(mvar) != stg_END_TSO_QUEUE_closure) {
1730
1731         /* There are takeMVar(s) waiting: wake up the first one
1732          */
1733         ASSERT(StgTSO_why_blocked(StgMVar_head(mvar)) == BlockedOnMVar::I16);
1734
1735         /* actually perform the takeMVar */
1736         tso = StgMVar_head(mvar);
1737         PerformTake(tso, R2);
1738         if (StgTSO_flags(tso) & TSO_DIRTY == 0) {
1739             foreign "C" dirty_TSO(MyCapability(), tso);
1740         }
1741
1742         ("ptr" tso) = foreign "C" unblockOne_(MyCapability() "ptr",
1743                                               StgMVar_head(mvar) "ptr", 1) [];
1744         StgMVar_head(mvar) = tso;
1745
1746         if (StgMVar_head(mvar) == stg_END_TSO_QUEUE_closure) {
1747             StgMVar_tail(mvar) = stg_END_TSO_QUEUE_closure;
1748         }
1749
1750 #if defined(THREADED_RTS)
1751         unlockClosure(mvar, stg_MVAR_DIRTY_info);
1752 #else
1753         SET_INFO(mvar,stg_MVAR_DIRTY_info);
1754 #endif
1755         jump %ENTRY_CODE(Sp(0));
1756     }
1757     else
1758     {
1759         /* No further takes, the MVar is now full. */
1760         StgMVar_value(mvar) = R2;
1761
1762 #if defined(THREADED_RTS)
1763         unlockClosure(mvar, stg_MVAR_DIRTY_info);
1764 #else
1765         SET_INFO(mvar,stg_MVAR_DIRTY_info);
1766 #endif
1767         jump %ENTRY_CODE(Sp(0));
1768     }
1769
1770     /* ToDo: yield afterward for better communication performance? */
1771 }
1772
1773
1774 tryPutMVarzh_fast
1775 {
1776     W_ mvar, info, tso;
1777
1778     /* args: R1 = MVar, R2 = value */
1779     mvar = R1;
1780
1781 #if defined(THREADED_RTS)
1782     ("ptr" info) = foreign "C" lockClosure(mvar "ptr") [R2];
1783 #else
1784     info = GET_INFO(mvar);
1785 #endif
1786
1787     if (StgMVar_value(mvar) != stg_END_TSO_QUEUE_closure) {
1788 #if defined(THREADED_RTS)
1789         unlockClosure(mvar, info);
1790 #endif
1791         RET_N(0);
1792     }
1793
1794     if (info == stg_MVAR_CLEAN_info) {
1795         foreign "C" dirty_MVAR(BaseReg "ptr", mvar "ptr");
1796     }
1797
1798     if (StgMVar_head(mvar) != stg_END_TSO_QUEUE_closure) {
1799
1800         /* There are takeMVar(s) waiting: wake up the first one
1801          */
1802         ASSERT(StgTSO_why_blocked(StgMVar_head(mvar)) == BlockedOnMVar::I16);
1803
1804         /* actually perform the takeMVar */
1805         tso = StgMVar_head(mvar);
1806         PerformTake(tso, R2);
1807         if (StgTSO_flags(tso) & TSO_DIRTY == 0) {
1808             foreign "C" dirty_TSO(MyCapability(), tso);
1809         }
1810
1811         ("ptr" tso) = foreign "C" unblockOne_(MyCapability() "ptr",
1812                                               StgMVar_head(mvar) "ptr", 1) [];
1813         StgMVar_head(mvar) = tso;
1814
1815         if (StgMVar_head(mvar) == stg_END_TSO_QUEUE_closure) {
1816             StgMVar_tail(mvar) = stg_END_TSO_QUEUE_closure;
1817         }
1818
1819 #if defined(THREADED_RTS)
1820         unlockClosure(mvar, stg_MVAR_DIRTY_info);
1821 #else
1822         SET_INFO(mvar,stg_MVAR_DIRTY_info);
1823 #endif
1824     }
1825     else
1826     {
1827         /* No further takes, the MVar is now full. */
1828         StgMVar_value(mvar) = R2;
1829
1830 #if defined(THREADED_RTS)
1831         unlockClosure(mvar, stg_MVAR_DIRTY_info);
1832 #else
1833         SET_INFO(mvar,stg_MVAR_DIRTY_info);
1834 #endif
1835     }
1836
1837     RET_N(1);
1838     /* ToDo: yield afterward for better communication performance? */
1839 }
1840
1841
1842 /* -----------------------------------------------------------------------------
1843    Stable pointer primitives
1844    -------------------------------------------------------------------------  */
1845
1846 makeStableNamezh_fast
1847 {
1848     W_ index, sn_obj;
1849
1850     ALLOC_PRIM( SIZEOF_StgStableName, R1_PTR, makeStableNamezh_fast );
1851
1852     (index) = foreign "C" lookupStableName(R1 "ptr") [];
1853
1854     /* Is there already a StableName for this heap object?
1855      *  stable_ptr_table is a pointer to an array of snEntry structs.
1856      */
1857     if ( snEntry_sn_obj(W_[stable_ptr_table] + index*SIZEOF_snEntry) == NULL ) {
1858         sn_obj = Hp - SIZEOF_StgStableName + WDS(1);
1859         SET_HDR(sn_obj, stg_STABLE_NAME_info, W_[CCCS]);
1860         StgStableName_sn(sn_obj) = index;
1861         snEntry_sn_obj(W_[stable_ptr_table] + index*SIZEOF_snEntry) = sn_obj;
1862     } else {
1863         sn_obj = snEntry_sn_obj(W_[stable_ptr_table] + index*SIZEOF_snEntry);
1864     }
1865
1866     RET_P(sn_obj);
1867 }
1868
1869
1870 makeStablePtrzh_fast
1871 {
1872     /* Args: R1 = a */
1873     W_ sp;
1874     MAYBE_GC(R1_PTR, makeStablePtrzh_fast);
1875     ("ptr" sp) = foreign "C" getStablePtr(R1 "ptr") [];
1876     RET_N(sp);
1877 }
1878
1879 deRefStablePtrzh_fast
1880 {
1881     /* Args: R1 = the stable ptr */
1882     W_ r, sp;
1883     sp = R1;
1884     r = snEntry_addr(W_[stable_ptr_table] + sp*SIZEOF_snEntry);
1885     RET_P(r);
1886 }
1887
1888 /* -----------------------------------------------------------------------------
1889    Bytecode object primitives
1890    -------------------------------------------------------------------------  */
1891
1892 newBCOzh_fast
1893 {
1894     /* R1 = instrs
1895        R2 = literals
1896        R3 = ptrs
1897        R4 = arity
1898        R5 = bitmap array
1899     */
1900     W_ bco, bitmap_arr, bytes, words;
1901
1902     bitmap_arr = R5;
1903
1904     words = BYTES_TO_WDS(SIZEOF_StgBCO) + StgArrWords_words(bitmap_arr);
1905     bytes = WDS(words);
1906
1907     ALLOC_PRIM( bytes, R1_PTR&R2_PTR&R3_PTR&R5_PTR, newBCOzh_fast );
1908
1909     bco = Hp - bytes + WDS(1);
1910     SET_HDR(bco, stg_BCO_info, W_[CCCS]);
1911
1912     StgBCO_instrs(bco)     = R1;
1913     StgBCO_literals(bco)   = R2;
1914     StgBCO_ptrs(bco)       = R3;
1915     StgBCO_arity(bco)      = HALF_W_(R4);
1916     StgBCO_size(bco)       = HALF_W_(words);
1917
1918     // Copy the arity/bitmap info into the BCO
1919     W_ i;
1920     i = 0;
1921 for:
1922     if (i < StgArrWords_words(bitmap_arr)) {
1923         StgBCO_bitmap(bco,i) = StgArrWords_payload(bitmap_arr,i);
1924         i = i + 1;
1925         goto for;
1926     }
1927
1928     RET_P(bco);
1929 }
1930
1931
1932 mkApUpd0zh_fast
1933 {
1934     // R1 = the BCO# for the AP
1935     //
1936     W_ ap;
1937
1938     // This function is *only* used to wrap zero-arity BCOs in an
1939     // updatable wrapper (see ByteCodeLink.lhs).  An AP thunk is always
1940     // saturated and always points directly to a FUN or BCO.
1941     ASSERT(%INFO_TYPE(%GET_STD_INFO(R1)) == HALF_W_(BCO) &&
1942            StgBCO_arity(R1) == HALF_W_(0));
1943
1944     HP_CHK_GEN_TICKY(SIZEOF_StgAP, R1_PTR, mkApUpd0zh_fast);
1945     TICK_ALLOC_UP_THK(0, 0);
1946     CCCS_ALLOC(SIZEOF_StgAP);
1947
1948     ap = Hp - SIZEOF_StgAP + WDS(1);
1949     SET_HDR(ap, stg_AP_info, W_[CCCS]);
1950
1951     StgAP_n_args(ap) = HALF_W_(0);
1952     StgAP_fun(ap) = R1;
1953
1954     RET_P(ap);
1955 }
1956
1957 unpackClosurezh_fast
1958 {
1959 /* args: R1 = closure to analyze */
1960 // TODO: Consider the absence of ptrs or nonptrs as a special case ?
1961
1962     W_ info, ptrs, nptrs, p, ptrs_arr, nptrs_arr;
1963     info  = %GET_STD_INFO(UNTAG(R1));
1964
1965     // Some closures have non-standard layout, so we omit those here.
1966     W_ type;
1967     type = TO_W_(%INFO_TYPE(info));
1968     switch [0 .. N_CLOSURE_TYPES] type {
1969     case THUNK_SELECTOR : {
1970         ptrs = 1;
1971         nptrs = 0;
1972         goto out;
1973     }
1974     case THUNK, THUNK_1_0, THUNK_0_1, THUNK_2_0, THUNK_1_1,
1975          THUNK_0_2, THUNK_STATIC, AP, PAP, AP_STACK, BCO : {
1976         ptrs = 0;
1977         nptrs = 0;
1978         goto out;
1979     }
1980     default: {
1981         ptrs  = TO_W_(%INFO_PTRS(info));
1982         nptrs = TO_W_(%INFO_NPTRS(info));
1983         goto out;
1984     }}
1985 out:
1986
1987     W_ ptrs_arr_sz, nptrs_arr_sz;
1988     nptrs_arr_sz = SIZEOF_StgArrWords   + WDS(nptrs);
1989     ptrs_arr_sz  = SIZEOF_StgMutArrPtrs + WDS(ptrs);
1990
1991     ALLOC_PRIM (ptrs_arr_sz + nptrs_arr_sz, R1_PTR, unpackClosurezh_fast);
1992
1993     W_ clos;
1994     clos = UNTAG(R1);
1995
1996     ptrs_arr  = Hp - nptrs_arr_sz - ptrs_arr_sz + WDS(1);
1997     nptrs_arr = Hp - nptrs_arr_sz + WDS(1);
1998
1999     SET_HDR(ptrs_arr, stg_MUT_ARR_PTRS_FROZEN_info, W_[CCCS]);
2000     StgMutArrPtrs_ptrs(ptrs_arr) = ptrs;
2001     p = 0;
2002 for:
2003     if(p < ptrs) {
2004          W_[ptrs_arr + SIZEOF_StgMutArrPtrs + WDS(p)] = StgClosure_payload(clos,p);
2005          p = p + 1;
2006          goto for;
2007     }
2008
2009     SET_HDR(nptrs_arr, stg_ARR_WORDS_info, W_[CCCS]);
2010     StgArrWords_words(nptrs_arr) = nptrs;
2011     p = 0;
2012 for2:
2013     if(p < nptrs) {
2014          W_[BYTE_ARR_CTS(nptrs_arr) + WDS(p)] = StgClosure_payload(clos, p+ptrs);
2015          p = p + 1;
2016          goto for2;
2017     }
2018     RET_NPP(info, ptrs_arr, nptrs_arr);
2019 }
2020
2021 /* -----------------------------------------------------------------------------
2022    Thread I/O blocking primitives
2023    -------------------------------------------------------------------------- */
2024
2025 /* Add a thread to the end of the blocked queue. (C-- version of the C
2026  * macro in Schedule.h).
2027  */
2028 #define APPEND_TO_BLOCKED_QUEUE(tso)                    \
2029     ASSERT(StgTSO__link(tso) == END_TSO_QUEUE);         \
2030     if (W_[blocked_queue_hd] == END_TSO_QUEUE) {        \
2031       W_[blocked_queue_hd] = tso;                       \
2032     } else {                                            \
2033       foreign "C" setTSOLink(MyCapability() "ptr", W_[blocked_queue_tl], tso); \
2034     }                                                   \
2035     W_[blocked_queue_tl] = tso;
2036
2037 waitReadzh_fast
2038 {
2039     /* args: R1 */
2040 #ifdef THREADED_RTS
2041     foreign "C" barf("waitRead# on threaded RTS") never returns;
2042 #else
2043
2044     ASSERT(StgTSO_why_blocked(CurrentTSO) == NotBlocked::I16);
2045     StgTSO_why_blocked(CurrentTSO) = BlockedOnRead::I16;
2046     StgTSO_block_info(CurrentTSO) = R1;
2047     // No locking - we're not going to use this interface in the
2048     // threaded RTS anyway.
2049     APPEND_TO_BLOCKED_QUEUE(CurrentTSO);
2050     jump stg_block_noregs;
2051 #endif
2052 }
2053
2054 waitWritezh_fast
2055 {
2056     /* args: R1 */
2057 #ifdef THREADED_RTS
2058     foreign "C" barf("waitWrite# on threaded RTS") never returns;
2059 #else
2060
2061     ASSERT(StgTSO_why_blocked(CurrentTSO) == NotBlocked::I16);
2062     StgTSO_why_blocked(CurrentTSO) = BlockedOnWrite::I16;
2063     StgTSO_block_info(CurrentTSO) = R1;
2064     // No locking - we're not going to use this interface in the
2065     // threaded RTS anyway.
2066     APPEND_TO_BLOCKED_QUEUE(CurrentTSO);
2067     jump stg_block_noregs;
2068 #endif
2069 }
2070
2071
2072 STRING(stg_delayzh_malloc_str, "delayzh_fast")
2073 delayzh_fast
2074 {
2075 #ifdef mingw32_HOST_OS
2076     W_ ares;
2077     CInt reqID;
2078 #else
2079     W_ t, prev, target;
2080 #endif
2081
2082 #ifdef THREADED_RTS
2083     foreign "C" barf("delay# on threaded RTS") never returns;
2084 #else
2085
2086     /* args: R1 (microsecond delay amount) */
2087     ASSERT(StgTSO_why_blocked(CurrentTSO) == NotBlocked::I16);
2088     StgTSO_why_blocked(CurrentTSO) = BlockedOnDelay::I16;
2089
2090 #ifdef mingw32_HOST_OS
2091
2092     /* could probably allocate this on the heap instead */
2093     ("ptr" ares) = foreign "C" stgMallocBytes(SIZEOF_StgAsyncIOResult,
2094                                             stg_delayzh_malloc_str);
2095     (reqID) = foreign "C" addDelayRequest(R1);
2096     StgAsyncIOResult_reqID(ares)   = reqID;
2097     StgAsyncIOResult_len(ares)     = 0;
2098     StgAsyncIOResult_errCode(ares) = 0;
2099     StgTSO_block_info(CurrentTSO)  = ares;
2100
2101     /* Having all async-blocked threads reside on the blocked_queue
2102      * simplifies matters, so change the status to OnDoProc put the
2103      * delayed thread on the blocked_queue.
2104      */
2105     StgTSO_why_blocked(CurrentTSO) = BlockedOnDoProc::I16;
2106     APPEND_TO_BLOCKED_QUEUE(CurrentTSO);
2107     jump stg_block_async_void;
2108
2109 #else
2110
2111     W_ time;
2112     W_ divisor;
2113     (time) = foreign "C" getourtimeofday() [R1];
2114     divisor = TO_W_(RtsFlags_MiscFlags_tickInterval(RtsFlags));
2115     if (divisor == 0) {
2116         divisor = 50;
2117     }
2118     divisor = divisor * 1000;
2119     target = ((R1 + divisor - 1) / divisor) /* divide rounding up */
2120            + time + 1; /* Add 1 as getourtimeofday rounds down */
2121     StgTSO_block_info(CurrentTSO) = target;
2122
2123     /* Insert the new thread in the sleeping queue. */
2124     prev = NULL;
2125     t = W_[sleeping_queue];
2126 while:
2127     if (t != END_TSO_QUEUE && StgTSO_block_info(t) < target) {
2128         prev = t;
2129         t = StgTSO__link(t);
2130         goto while;
2131     }
2132
2133     StgTSO__link(CurrentTSO) = t;
2134     if (prev == NULL) {
2135         W_[sleeping_queue] = CurrentTSO;
2136     } else {
2137         foreign "C" setTSOLink(MyCapability() "ptr", prev, CurrentTSO) [];
2138     }
2139     jump stg_block_noregs;
2140 #endif
2141 #endif /* !THREADED_RTS */
2142 }
2143
2144
2145 #ifdef mingw32_HOST_OS
2146 STRING(stg_asyncReadzh_malloc_str, "asyncReadzh_fast")
2147 asyncReadzh_fast
2148 {
2149     W_ ares;
2150     CInt reqID;
2151
2152 #ifdef THREADED_RTS
2153     foreign "C" barf("asyncRead# on threaded RTS") never returns;
2154 #else
2155
2156     /* args: R1 = fd, R2 = isSock, R3 = len, R4 = buf */
2157     ASSERT(StgTSO_why_blocked(CurrentTSO) == NotBlocked::I16);
2158     StgTSO_why_blocked(CurrentTSO) = BlockedOnRead::I16;
2159
2160     /* could probably allocate this on the heap instead */
2161     ("ptr" ares) = foreign "C" stgMallocBytes(SIZEOF_StgAsyncIOResult,
2162                                             stg_asyncReadzh_malloc_str)
2163                         [R1,R2,R3,R4];
2164     (reqID) = foreign "C" addIORequest(R1, 0/*FALSE*/,R2,R3,R4 "ptr") [];
2165     StgAsyncIOResult_reqID(ares)   = reqID;
2166     StgAsyncIOResult_len(ares)     = 0;
2167     StgAsyncIOResult_errCode(ares) = 0;
2168     StgTSO_block_info(CurrentTSO)  = ares;
2169     APPEND_TO_BLOCKED_QUEUE(CurrentTSO);
2170     jump stg_block_async;
2171 #endif
2172 }
2173
2174 STRING(stg_asyncWritezh_malloc_str, "asyncWritezh_fast")
2175 asyncWritezh_fast
2176 {
2177     W_ ares;
2178     CInt reqID;
2179
2180 #ifdef THREADED_RTS
2181     foreign "C" barf("asyncWrite# on threaded RTS") never returns;
2182 #else
2183
2184     /* args: R1 = fd, R2 = isSock, R3 = len, R4 = buf */
2185     ASSERT(StgTSO_why_blocked(CurrentTSO) == NotBlocked::I16);
2186     StgTSO_why_blocked(CurrentTSO) = BlockedOnWrite::I16;
2187
2188     ("ptr" ares) = foreign "C" stgMallocBytes(SIZEOF_StgAsyncIOResult,
2189                                             stg_asyncWritezh_malloc_str)
2190                         [R1,R2,R3,R4];
2191     (reqID) = foreign "C" addIORequest(R1, 1/*TRUE*/,R2,R3,R4 "ptr") [];
2192
2193     StgAsyncIOResult_reqID(ares)   = reqID;
2194     StgAsyncIOResult_len(ares)     = 0;
2195     StgAsyncIOResult_errCode(ares) = 0;
2196     StgTSO_block_info(CurrentTSO)  = ares;
2197     APPEND_TO_BLOCKED_QUEUE(CurrentTSO);
2198     jump stg_block_async;
2199 #endif
2200 }
2201
2202 STRING(stg_asyncDoProczh_malloc_str, "asyncDoProczh_fast")
2203 asyncDoProczh_fast
2204 {
2205     W_ ares;
2206     CInt reqID;
2207
2208 #ifdef THREADED_RTS
2209     foreign "C" barf("asyncDoProc# on threaded RTS") never returns;
2210 #else
2211
2212     /* args: R1 = proc, R2 = param */
2213     ASSERT(StgTSO_why_blocked(CurrentTSO) == NotBlocked::I16);
2214     StgTSO_why_blocked(CurrentTSO) = BlockedOnDoProc::I16;
2215
2216     /* could probably allocate this on the heap instead */
2217     ("ptr" ares) = foreign "C" stgMallocBytes(SIZEOF_StgAsyncIOResult,
2218                                             stg_asyncDoProczh_malloc_str)
2219                                 [R1,R2];
2220     (reqID) = foreign "C" addDoProcRequest(R1 "ptr",R2 "ptr") [];
2221     StgAsyncIOResult_reqID(ares)   = reqID;
2222     StgAsyncIOResult_len(ares)     = 0;
2223     StgAsyncIOResult_errCode(ares) = 0;
2224     StgTSO_block_info(CurrentTSO) = ares;
2225     APPEND_TO_BLOCKED_QUEUE(CurrentTSO);
2226     jump stg_block_async;
2227 #endif
2228 }
2229 #endif
2230
2231 // noDuplicate# tries to ensure that none of the thunks under
2232 // evaluation by the current thread are also under evaluation by
2233 // another thread.  It relies on *both* threads doing noDuplicate#;
2234 // the second one will get blocked if they are duplicating some work.
2235 noDuplicatezh_fast
2236 {
2237     SAVE_THREAD_STATE();
2238     ASSERT(StgTSO_what_next(CurrentTSO) == ThreadRunGHC::I16);
2239     foreign "C" threadPaused (MyCapability() "ptr", CurrentTSO "ptr") [];
2240
2241     if (StgTSO_what_next(CurrentTSO) == ThreadKilled::I16) {
2242         jump stg_threadFinished;
2243     } else {
2244         LOAD_THREAD_STATE();
2245         ASSERT(StgTSO_what_next(CurrentTSO) == ThreadRunGHC::I16);
2246         jump %ENTRY_CODE(Sp(0));
2247     }
2248 }
2249
2250 getApStackValzh_fast
2251 {
2252    W_ ap_stack, offset, val, ok;
2253
2254    /* args: R1 = AP_STACK, R2 = offset */
2255    ap_stack = R1;
2256    offset   = R2;
2257
2258    if (%INFO_PTR(ap_stack) == stg_AP_STACK_info) {
2259         ok = 1;
2260         val = StgAP_STACK_payload(ap_stack,offset);
2261    } else {
2262         ok = 0;
2263         val = R1;
2264    }
2265    RET_NP(ok,val);
2266 }